A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System...
Transcript of A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System...
![Page 1: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/1.jpg)
A Secure Collaboration System for Coal
Supply Chains in Australia
Shiping Chen and Chen Wang
Information Engineering Lab
CSIRO ICT Centre, Australia
Microsoft Cloud Futures 2011, June 3
![Page 2: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/2.jpg)
Outlines
• Background of Coal Chains
• Challenge for Collaboration in Coal Chains
• Information Flow Control Mechanism
• The Role of “Cloud” in Information Flow Control
• Conclusion
![Page 3: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/3.jpg)
Coal Supply Chain Example
![Page 4: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/4.jpg)
Coal Supply Chain Example
40 Mines owned by
13 coal producers
![Page 5: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/5.jpg)
Coal Supply Chain Example
3 large haulage operators
2 track owner/operators
![Page 6: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/6.jpg)
Coal Supply Chain Example
6 coal dumping stations
7 ship berths
![Page 7: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/7.jpg)
The Need of Coordination in Coal Chain
• A typical transport supply chain in Australian coal industry
involves multiple business entities that own different resources
• Hunter Valley Coal Chain: 40 coal mines owned by 13 producers,
27 load points, 28 trains run by 3 rail operators, tracks own by 2
operators, 3 coal loading terminals, 7 ship berths, 9 vessel agents,
34 end buyers from 12 countries…
• The producers and operators are independent organizations
• They contract with each other to ensure the resources for shipping
the coal.
• Owners of same type of resources compete with each other.
• Problems
• Individually negotiated contracts may not lead to an optimal (or
sometimes even feasible) resource usage in the whole coal chain.
![Page 8: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/8.jpg)
The Need of Coordination in Coal Chain
• Two vessels ship different brands of coals from the same berth
• V1: (100 tones: mine A, 100 tones: mine B); arrives at 9:30
• V2: (200 tones: mine C, 100 tones: mine D); arrives at 10:30
• Miners negotiate with rail operators independently for shipping
coals to the berth
• Miner A, D negotiate with Rail Operator r1
• Miner B, C negotiate with Rail Operator r2
• Rail Operator r1 and r2 contract with track operator T for track
allocation
![Page 9: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/9.jpg)
The Need of Coordination in Coal Chain
• Schedule 1
• Vessel waiting time: v1 – 1 hour; v2 – 0.5 hour
• Schedule 2
• Vessel waiting time: v1 – 0 hour; v2 – 0 hour
![Page 10: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/10.jpg)
• The coal chain can run more efficiently with a coordinator
• The obstacle of setting up a coordinator
• miners or operators need to make some sensitive information available to
the coordinator,
• e.g., Miner A needs to disclose to the coordinator that it needs to transport 100
tones coal to stockyard 1 for loading to vessel v1 that arrives at 9:30.
• The current coal chain practices do not guarantee that the information
won’t flow to the competitors of the information provider.
• Convincing miners and operators to share information is difficult.
Challenges for Sharing Data in Coal Chain
![Page 11: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/11.jpg)
System Requirement
• The system should ensure that shared information is only used
for the purpose specified by the information provider
• The system should allow the party that supplies the data to specify
who can use the data and how the data should be used;
• The owner specified policies should be enforced when data flow
across administrative boundaries;
• A user-supplied program that accesses a set of data in the system
should label information flow inside the program
• For checking if the program satisfies the access control and information
flow policies of its data owner.
• The access to the output produced by an application from a set of data
should also satisfy certain policies specified by the owners of input
data.
![Page 12: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/12.jpg)
Information Flow Control
• The system shall consist of the following components in order
to meet the requirements:
• A secure data store
• Only allows authorized parties to access data according to the access
control policies specified by the data owner.
• Enforces information flow control policies.
• A secure execution environment
• Provides a mechanism to ensure that user-supplied programs follow
information flow policies associated with data they access.
• Provides isolation mechanism for running user-supplied programs..
![Page 13: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/13.jpg)
Information Flow Control: Example
Secure data store
Secure execution environment
Rail operator
scheduler(r)
Miner A
Port operator
scheduler
Track
operator
scheduler
1. PutData(d)
2.Authorize(r,d)
3.GetData(d)
(A, x, v1, 9:30, s1)
(A, x, 8:30, 9:30, r1)
4.PutData(s(d))
2.Authorize(b,s(d))
5.GetData(s)
Miner B
![Page 14: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/14.jpg)
LTL based Data Labelling
• Request is not allowed to propagate to a party without the
approval from the owner
• Source = s ^ Request = m ^ Target = t ^ (¬(r t) U Approve(s, t,
m))
• A scheduler cannot process a request without approval from
the owner
• Source = s ^ Request = m ^ Scheduler= c ^ (¬c.schedule(m) U
Approve(s, c, r))
• The scheduler output cannot flow to a party without the
approval from the owner
• Source = s ^ Request = m ^ Scheduler= c ^ Target = t ^
(¬c.schedule(m) t U Approve(s, c.schedule(m), t))
![Page 15: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/15.jpg)
Rail operator – submit code
![Page 16: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/16.jpg)
Miner – review the code accessing the data
![Page 17: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/17.jpg)
Miner – submit and label request
![Page 18: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/18.jpg)
Limitation of Information Flow Control
• Who plays the coordinator’s role?
• It is difficult to setup an independent party for the coordinating
process.
• It is technically challenging to manage and maintain a secure
execution environment and a secure data store.
• Threats
• Software bugs
• Inside attack: the coordinator is formed by stakeholders
• Tamper execution environment
• Leak data from the store
![Page 19: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/19.jpg)
Accountability
“People think that security in the real world is based on locks. In
fact, realworld security depends mainly on deterrence, and
hence on the possibility of punishment.”
Butler Lampson: “Privacy and security - Usable security: how to get it.”
CACM 52(11), 2009
“Accountability is the ability to hold an entity, such as a person or
organization, responsible for its actions.”
Butler Lampson:“Accountability and Freedom”, 2005
![Page 20: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/20.jpg)
The Role of “Cloud”
• Accountability in business world
• The use of a trusted third party to make a deal
• The use of legal/social systems
• Contract law provides incentives that promote good behaviour between
parties
• Using the “cloud” as a middleman if the cloud provider is more
trustworthy for a party than its collaborative parties
• Execution environment of each party is isolated in the cloud.
• Data and program labels describing how information may flow is
visible by the “cloud”.
• A party can be caught accountable by the “cloud” when violating the
information flow policy.
• The middleman’s role
• Evidence collection based on disclosed policies associated with data
and programs
• Runtime compliance check and problem detection
![Page 21: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/21.jpg)
Coal Supply Chain in the Cloud
cloud
Evidence store
Online
log
analyser
source | dest | label | data feature
![Page 22: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/22.jpg)
Preliminary Performance Evaluation
• Amazon EC2 small instances
• Simple storage service (put/get)
• Tomcat + axis2 + BerkeleyDB XML
• Logging overhead
0
200
400
600
800
1000
1200
1400
1600
0.1 1 10 100 1000
Th
rou
gh
pu
t(K
B/s
)
Average data size (KB/request)
Throughput Comparison
w/o logging
w logging
![Page 23: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/23.jpg)
Preliminary Performance Evaluation
• SOAP Message Reconstruction Cost
0
100
200
300
400
500
600
700
800
0.1 1 10 100 1000
Av
era
ge
Pro
ce
ss
ing
Tim
e(m
s)
Request Size (KB)
Simple Log Proc. Time vs. Request Proc. Time
log processing time
request proc time
![Page 24: A Secure Collaboration System for Coal Supply Chains in ...€¦ · A Secure Collaboration System for Coal Supply Chains in Australia Shiping Chen and Chen Wang Information Engineering](https://reader033.fdocuments.in/reader033/viewer/2022060520/604eba17ee66447b6f4450e3/html5/thumbnails/24.jpg)
Conclusion
• Data sharing in business collaborations is difficult even for
achieving common benefit
• Collaborating parties often competing with each other at the same
time
• Information flow control alone cannot address this problem
• The cloud computing paradigm offers opportunities to solve the
problem
• Cloud computing technologies provide effective isolation for data
and compute of collaborating parties.
• An independent cloud infrastructure provider creates a
middleman’s role even though cloud platform itself has
trustworthiness problem
• capable of collecting evidences based on interactions between
collaborating parties.